Search

ABSTRACT Recently, significant progress has been made in completing static knowledge graphs. However, knowledge tends to evolve with time, and static knowledge graph completion (KGC) methods struggle to capture the changes. Therefore, temporal knowledge graph (TKG) reasoning has become a focus of research. Most existing TKG methods incorporate temporal information into triplets and transform them into KGC tasks, ignoring the important influence of time information and implicit relationships between entities. In this paper, we propose a new method called TD‐RKG, which addresses the challenges of temporal variability and implicit entity correlations based on a dynamic fusion representation learning approach. The method consists of four modules: dynamic local recurrent encoding layer, dynamic implicit encoding layer, dynamic global information attention layer and decoding layer. Experimental results on three benchmark datasets demonstrate substantial improvements in TD‐RKG across multiple evaluation metrics. [ABSTRACT FROM AUTHOR]

Effective use of the wealth of information provided by Distributed Acoustic Sensing (DAS) for mass movement monitoring remains a challenge. We propose a semi‐supervised neural network tailored to screen DAS data related to a series of rock collapses leading to a major failure of approximately 1.2 million m3 ${\mathrm{m}}^{3}$ on 15 June 2023 in Brienz, Eastern Switzerland. Besides DAS, the dataset from 16 May to 30 June 2023 includes Doppler radar data for partially ground‐truth labeling. The proposed algorithm is capable of distinguishing between rock‐slope failures and background noise, including road and train traffic, with a detection precision of over 95% $95\%$. It identifies hundreds of precursory failures and shows sustained detection hours before and during the major collapse. Event size and signal‐to‐noise ratio (SNR) are the key performance dependencies. As a critical part of our algorithm operates unsupervised, we suggest that it is suitable for general monitoring of natural hazards. Plain Language Summary: Steep mountains and hills produce dangerous rockfalls and similar phenomena such as landslides and debris flows. A major collapse is typically preceded by a series of rockfalls over days or months. It is therefore crucial to reliably detect these events and recognize the warning signs of an impending major collapse. When rocks bounce on the ground they release seismic waves, which generate vibrations that propagate long distances. Such vibrations stretch and compress fiber optic cables within the ground enough so they can be measured with a novel technique called Distributed Acoustic Sensing (DAS). Here we show how to identify such DAS signals using machine learning algorithms to detect precursory rockfall activity and a major collapse on a slope in Switzerland. We compare our detections with radar measurements, which are highly reliable but also come at a greater cost for installation. Since we can apply DAS to unused fiber within the ground, our approach may pave the way for the next generation of natural hazard warning. Key Points: A semi‐supervised neural network is developed for rock‐slope failure monitoring with Distributed Acoustic Sensing at Brienz, SwitzerlandOur model achieves over 95% precision for rock slope failures detected by a Doppler radar system over 45 daysThe sustained detection of slope failures before the major collapse highlights the potential of our approach for early warning [ABSTRACT FROM AUTHOR]

Background: The rapid advancements in deep neural network models have significantly enhanced the ability to extract features from microbial sequence data, which is critical for addressing biological challenges. However, the scarcity and complexity of labeled microbial data pose substantial difficulties for supervised learning approaches. To address these issues, we propose DNASimCLR, an unsupervised framework designed for efficient gene sequence data feature extraction. Results: DNASimCLR leverages convolutional neural networks and the SimCLR framework, based on contrastive learning, to extract intricate features from diverse microbial gene sequences. Pre-training was conducted on two classic large scale unlabelled datasets encompassing metagenomes and viral gene sequences. Subsequent classification tasks were performed by fine-tuning the pretrained model using the previously acquired model. Our experiments demonstrate that DNASimCLR is at least comparable to state-of-the-art techniques for gene sequence classification. For convolutional neural network-based approaches, DNASimCLR surpasses the latest existing methods, clearly establishing its superiority over the state-of-the-art CNN-based feature extraction techniques. Furthermore, the model exhibits superior performance across diverse tasks in analyzing biological sequence data, showcasing its robust adaptability. Conclusions: DNASimCLR represents a robust and database-agnostic solution for gene sequence classification. Its versatility allows it to perform well in scenarios involving novel or previously unseen gene sequences, making it a valuable tool for diverse applications in genomics. [ABSTRACT FROM AUTHOR]

Bitter taste is an unpleasant taste modality that affects food consumption. Bitter peptides are generated during enzymatic processes that produce functional, bioactive protein hydrolysates or during the aging process of fermented products such as cheese, soybean protein, and wine. Understanding the underlying peptide sequences responsible for bitter taste can pave the way for more efficient identification of these peptides. This paper presents BitterPep-GCN, a feature-agnostic graph convolution network for bitter peptide prediction. The graph-based model learns the embedding of amino acids in the bitter peptide sequences and uses mixed pooling for bitter classification. BitterPep-GCN was benchmarked using BTP640, a publicly available bitter peptide dataset. The latent peptide embeddings generated by the trained model were used to analyze the activity of sequence motifs responsible for the bitter taste of the peptides. Particularly, we calculated the activity for individual amino acids and dipeptide, tripeptide, and tetrapeptide sequence motifs present in the peptides. Our analyses pinpoint specific amino acids, such as F, G, P, and R, as well as sequence motifs, notably tripeptide and tetrapeptide motifs containing FF, as key bitter signatures in peptides. This work not only provides a new predictor of bitter taste for a more efficient identification of bitter peptides in various food products but also gives a hint into the molecular basis of bitterness. Scientific Contribution Our work provides the first application of Graph Neural Networks for the prediction of peptide bitter taste. The best-developed model, BitterPep-GCN, learns the embedding of amino acids in the bitter peptide sequences and uses mixed pooling for bitter classification. The embeddings were used to analyze the sequence motifs responsible for the bitter taste. [ABSTRACT FROM AUTHOR]

Networks of interconnected neurons communicating through spiking signals offer the bedrock of neural computations. Our brain's spiking neural networks have the computational capacity to achieve complex pattern recognition and cognitive functions effortlessly. However, solving real-world problems with artificial spiking neural networks (SNNs) has proved to be difficult for a variety of reasons. Crucially, scaling SNNs to large networks and processing large-scale real-world datasets have been challenging, especially when compared to their non-spiking deep learning counterparts. The critical operation that is needed of SNNs is the ability to learn distributed representations from data and use these representations for perceptual, cognitive and memory operations. In this work, we introduce a novel SNN that performs unsupervised representation learning and associative memory operations leveraging Hebbian synaptic and activity- dependent structural plasticity coupled with neuron-units modelled as Poisson spike generators with sparse firing (~1Hz mean and ~100 Hz maximum firing rate). Crucially, the architecture of our model derives from the neocortical columnar organization and combines feed forward projections for learning hidden representations and recurrent projections for forming associative memories. We evaluated the model on properties relevant for attractor-based associative memories such as pattern completion, perceptual rivalry, distortion resistance, and prototype extraction. [ABSTRACT FROM AUTHOR]

Often, applications of self-supervised learning to 3D medical data opt to use 3D variants of successful 2D network architectures. Although promising approaches, they are significantly more computationally demanding to train, and thus reduce the widespread applicability of these methods away from those with modest computational resources. Thus, in this paper, we aim to improve standard 2D SSL algorithms by modelling the inherent 3D nature of these datasets implicitly. We propose two variants that build upon a strong baseline model and show that both of these variants often outperform the baseline in a variety of downstream tasks. Importantly, in contrast to previous works in both 2D and 3D approaches for 3D medical data, both of our proposals introduce negligible additional overhead in terms of parameter complexity. Although data loading overhead increases over the baseline SimCLR model (which we can show can be somewhat mitigated through parallelisation), our proposed models are still significantly more efficient than previous approaches based on sequence modelling. Overall, our proposed methods help improve the democratisation of these approaches for medical applications. • Introduce 2 models that implicitly model 3D pathology data and improve performance. • Conduct detailed ablations of the hyperparameters and design choices for our methods. • A extensive evaluation covering multiple downstream datasets, tasks, and modalities. [ABSTRACT FROM AUTHOR]

Collaborative Filtering (CF) plays a crucial role in modern recommender systems, leveraging historical user-item interactions to provide personalized suggestions. However, CF-based methods often encounter biases due to imbalances in training data. This phenomenon makes CF-based methods tend to prioritize recommending popular items and performing unsatisfactorily on inactive users. Existing works address this issue by rebalancing training samples, reranking recommendation results, or making the modeling process robust to the bias. Despite their effectiveness, these approaches can compromise accuracy or be sensitive to weighting strategies, making them challenging to train. Therefore, exploring how to mitigate these biases remains in urgent demand. In this article, we deeply analyze the causes and effects of the biases and propose a framework to alleviate biases in recommendation from the perspective of representation distribution, namely Group-Alignment and Global-Uniformity Enhanced Representation Learning for Debiasing Recommendation (AURL). Specifically, we identify two significant problems in the representation distribution of users and items, namely group-discrepancy and global-collapse. These two problems directly lead to biases in the recommendation results. To this end, we propose two simple but effective regularizers in the representation space, respectively named group-alignment and global-uniformity. The goal of group-alignment is to bring the representation distribution of long-tail entities closer to that of popular entities, while global-uniformity aims to preserve the information of entities as much as possible by evenly distributing representations. Our method directly optimizes both the group-alignment and global-uniformity regularization terms to mitigate recommendation biases. Please note that AURL applies to arbitrary CF-based recommendation backbones. Extensive experiments on three real datasets and various recommendation backbones verify the superiority of our proposed framework. The results show that AURL not only outperforms existing debiasing models in mitigating biases but also improves recommendation performance to some extent. [ABSTRACT FROM AUTHOR]

Representation learning is an important topic in machine learning, pattern recognition, and data mining research. Among many representation learning approaches, semi-nonnegative matrix factorization (SNMF) is a frequently-used one. However, a typical problem of SNMF is that usually there is no learning rate guidance during the optimization process, which often leads to a poor representation ability. To overcome this limitation, we propose a very general representation learning framework (DNSRF) that is based on a deep neural net. Essentially, the parameters of the deep net used to construct the DNSRF algorithms are obtained by matrix element update. In combination with different activation functions, DNSRF can be implemented in various ways. In our experiments, we tested nine instances of our DNSRF framework on six benchmark datasets. In comparison with other state-of-the-art methods, the results demonstrate the superior performance of our framework, which is thus shown to have a great representation ability. [ABSTRACT FROM AUTHOR]

Dynamic latent scale GAN is an architecture‐agnostic encoder‐based generative model inversion method. This paper introduces a method to efficiently integrate perceptual VAE into dynamic latent scale GAN to improve the performance of dynamic latent scale GAN. When dynamic latent scale GAN is trained with a normal i.i.d. latent random variable and the latent encoder is integrated into the discriminator, a sum of a predicted latent random variable of real data and a scaled normal noise follows the normal i.i.d. random variable. Since this random variable is paired with real data and follows the latent random variable, it can be used for both VAE and GAN training. Furthermore, by considering the intermediate layer output of the discriminator as the feature encoder output, the VAE can be trained to minimise the perceptual reconstruction loss. The forward propagation & backpropagation for minimising this perceptual reconstruction loss can be integrated with those of GAN training. Therefore, the proposed method does not require additional computations compared to typical GAN or dynamic latent scale GAN. Integrating perceptual VAE to dynamic latent scale GAN improved the generative and inversion performance of the model. [ABSTRACT FROM AUTHOR]

Hyperspectral small target detection (HSTD) is a promising pixel-level detection task. However, due to the low contrast and imbalanced number between the target and the background spatially and the high dimensions spectrally, it is a challenging one. To address these issues, this work proposes a representation-learning-based graph and generative network for hyperspectral small target detection. The model builds a fusion network through frequency representation for HSTD, where the novel architecture incorporates irregular topological data and spatial–spectral features to improve its representation ability. Firstly, a Graph Convolutional Network (GCN) module better models the non-local topological relationship between samples to represent the hyperspectral scene's underlying data structure. The mini-batch-training pattern of the GCN decreases the high computational cost of building an adjacency matrix for high-dimensional data sets. In parallel, the generative model enhances the differentiation reconstruction and the deep feature representation ability with respect to the target spectral signature. Finally, a fusion module compensates for the extracted different types of HS features and integrates their complementary merits for hyperspectral data interpretation while increasing the detection and background suppression capabilities. The performance of the proposed approach is evaluated using the average scores of AU C D , F , AU C F , τ , AU C BS , and AU C SNPR . The corresponding values are 0.99660, 0.00078, 0.99587, and 333.629, respectively. These results demonstrate the accuracy of the model in different evaluation metrics, with AU C D , F achieving the highest score, indicating strong detection performance across varying thresholds. Experiments on different hyperspectral data sets demonstrate the advantages of the proposed architecture. [ABSTRACT FROM AUTHOR]

Modern Multi-Function Radars (MFRs) are sophisticated sensors that are capable of flexibly adapting their control parameters in transmitted pulse sequences. In complex electromagnetic environments, efficiently and accurately recognizing the inter-pulse modulations of non-cooperative radar pulse sequences is a key step for modern Electronic Support (ES) systems. Existing recognition methods focus more on algorithmic designs, such as neural network structure designs, to improve recognition performance. However, in open electromagnetic environments with increased flexibility in radar transmission, these methods would suffer performance degradation due to domain shifts between training and testing datasets. To address this issue, this study proposes a robust radar inter-pulse modulation feature extraction and recognition method based on disentangled representation learning. At first, inspired by the Representation Learning Theory (RLT), the received radar pulse sequences can be disentangled into three explanatory factors related to (i) modulation types, (ii) modulation parameters, and (iii) measurement characteristics, such as measurement noise. Then, an explainable radar pulse sequence disentanglement network is proposed based on auto-encoding variational Bayes. The features extracted through the proposed method can effectively represent the key latent factors related to recognition tasks and maintain performance under domain shift conditions. Experiments on both ideal and non-ideal situations demonstrate the effectiveness, robustness, and superiority of the proposed method in comparison with other methods. [ABSTRACT FROM AUTHOR]

Recent advancements in deep neural networks have shown great potential in generating realistic data and performing clustering tasks. This is due to their ability to capture intricate patterns. However, current generative models face challenges such as poor performance and computational complexity caused by the issue of dimension disaster. The variational autoencoder (VAE), a commonly used method, also encounters problems such as posterior collapse and poor performance in multiclass classification when using the latent variables of VAE. Our goal in this study is to tackle the issue of effective disentanglement in image generation, classification and clustering tasks. We develop a generative network based on VAE incorporating a Gaussian mixture distribution as the prior. This enhancement improves the representation of latent variables and helps to overcome the challenges of matching the ground truth posterior. To further improve clustering performance, we introduce the total correlation as a kernel for computing latent features between embedding points and cluster centers. This technique is particularly useful in cases with complex latent variables and can also be applied for hierarchical disentanglement. Moreover, we employ the Fisher discriminant as a regularization term to minimize the within-class distance and maximize the between-class distance for samples, which has an important effect on the performance of our model viewed from the experimental results. We evaluate our proposed network on four datasets, and the experimental results demonstrate its effectiveness across multiple metrics. [ABSTRACT FROM AUTHOR]

The deployment of deep learning architectures on low-computational resource devices is challenging due to their high number of parameters and computational complexity. These heavy and complex architectures result in increased latency in real-time applications. However, splitting the deep architecture in a way that parallelizes the forward propagation into different subnets deploying into multiple low-computational resource devices, and then, aggregating the predictions may reduce the latency while preserving the performance. In this paper, we propose a novel deep learning architecture called Ensembled Parallel Networks (EnParaNets) that leverage network dissection, knowledge distillation, and ensemble learning strategies to reduce inference time while maintaining, even in some cases, outperforming the baseline accuracy in real-time applications. The methodology involves splitting the original network into N equal-sized blocks, forming N Sub-ParaNets for each block, and enhancing their representations using (A) contrastive knowledge distillation along with reducing Kullback–Leibler divergence between logits distributions of the teacher and student networks, and (B) L2 loss between intermediate representations of the original network and corresponding Sub-ParaNets. Predictive distributions from each Sub-ParaNet are assembled to form the final EnParaNet. The proposed EnParaNet outperforms the baseline models of seven diverse architectures: ResNet56, VGG_13, WRN_40_2, DenseNet, ResNeXt50, MobileNetv2, and ShuffleNetv2 in terms of accuracy while reducing inference time significantly using training methods A and B, respectively. Our proposed EnParaNet enhances ResNet56, VGG_13, WRN_40_2, MobileNetv2, DenseNet, ResNeXt50, and ShuffleNetv2 by 2.69%, 0.24%, 1.95%, 7.69%, 0.33%, 2.13%, and 3.12%, respectively, using training method A, where the inference time is reduced by 45%, 24%, 47%, 31%, 33%, 32%, and 44%, respectively. With training method B, EnParaNet achieves improvements of 1.75%, 2.90%, 1.09%, 3.91%, and 1.66%, with inference time reductions of 50%, 42%, 49%, 48%, and 49%, respectively. Moreover, a comprehensive ablation study analyzes the performance of the proposed technique and highlights its effectiveness and challenges. Furthermore, we also evaluate the performance of EnParaNet in transferability and adversarial robustness tasks. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer-Aided Design & Computer Graphics / Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao is the property of Gai Kan Bian Wei Hui and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Deep reinforcement learning has significantly advanced robot manipulations by providing an alternative solution for designing control strategies using raw images as direct inputs. While images offer additional environmental information, the end-to-end policy training manner (from image to action) requires simultaneous representation and task learning by the agent. This often necessitates a substantial number of interaction samples to achieve satisfactory policy performance. Previous works has attempted to address this challenge by learning a visual representation model that encodes the entire image into a low-dimensional vector before the policy training. However, since this vector contains both robot and object information, it inevitably introduces coupling within the state, which can mislead the policy training process. In this study, a novel method called Reinforcement Learning with Decoupled State Representation is proposed to effectively decouple robot and object information within the state representation. Experimental results demonstrate that the proposed method exhibits faster learning speed and achieves superior performance compared to previous methods across various robot manipulation tasks. Moreover, with only 3096 offline images, the proposed method successfully applies to real-world robot pushing tasks, which demonstrates its high practicability. [ABSTRACT FROM AUTHOR]

This research addresses the enhancement of face anti-spoofing (FAS) in facial recognition systems (FRS) against sophisticated fraudulent activities. Prior methods primarily focus on extracting facial features like color, texture, and dynamic variations, yet these methods struggle to accurately identify common characteristics of forged faces, thereby limiting generalization in practical scenarios. This study aims to propose a novel representation learning framework incorporating adversarial learning algorithms, to segregate features into liveness-specific and domain-specific categories, emphasizing liveness-specific features for FAS advancement. Feature disentanglement is central to this approach, enabling the deep learning models to effectively discern separable latent generating factors, such as identity, liveness, appearance, and texture. This methodology enhances model interpretability, explainability, and generalization. Additionally, Grad-CAM is employed to elucidate the basis for classifications made by the architecture, increasing explainability and trustworthiness. Empirical evaluation across panoply available FAS datasets confirms the superiority, significantly improving performance and robustness over existing technologies. [ABSTRACT FROM AUTHOR]

Dynamic Graph Neural Network (DGNN) models have been widely used for modelling, prediction and recommendation tasks in domains such as e-commerce and social networks, due to their ability to capture node interaction features and temporal features. Current methods for dynamic graph representation learning mainly depend on querying K-hop neighbors and the triadic closure law to derive node representations. However, as the number of layers in neural networks increases, this can cause problems with over-smoothing and overly complex calculations. Additionally, current models cannot ensure that events arrive at adjacent nodes in chronological order according to timestamps. To address these problems, we propose a Dynamic Graph Neural Network via Memory Regenerate and Neighbor Propagation(DGNN-MN) model. The model presents a memory regeneration strategy for obtaining node time information features and a time edge-propagating method for obtaining neighbour information. By combining these two methods to fuse output vectors, it captures node feature representations. In addition, we present a strategy for the timestamp encoding of node messages, which effectively ensures that node messages propagate to neighboring nodes in an ordered manner according to timestamps, thereby better capturing the temporal characteristics of events in dynamic graphs. Extensive experiments conducted on five public datasets demonstrate the effectiveness of DGNN-MN for link prediction and node classification task. Furthermore, the method outperforms other state-of-the-art methods. The data and code are available on GitHub. [ABSTRACT FROM AUTHOR]

The prevalence of long-tailed distributions in real-world data often results in classification models favoring the dominant classes, neglecting the less frequent ones. Current approaches address the issues in long-tailed image classification by rebalancing data, optimizing weights, and augmenting information. However, these methods often struggle to balance the performance between dominant and minority classes because of inadequate representation learning of the latter. To address these problems, we introduce descriptional words into images as cross-modal privileged information and propose a cross-modal enhanced method for long-tailed image classification, referred to as CMLTNet. CMLTNet improves the learning of intraclass similarity of tail-class representations by cross-modal alignment and captures the difference between the head and tail classes in semantic space by cross-modal inference. After fusing the above information, CMLTNet achieved an overall performance that was better than those of benchmark long-tailed and cross-modal learning methods on the long-tailed cross-modal datasets, NUS-WIDE and VireoFood-172. The effectiveness of the proposed modules was further studied through ablation experiments. In a case study of feature distribution, the proposed model was better in learning representations of tail classes, and in the experiments on model attention, CMLTNet has the potential to help learn some rare concepts in the tail class through mapping to the semantic space. [ABSTRACT FROM AUTHOR]

Micro-expression (ME) is an instinctive reaction that is not controlled by thoughts. It reveals one's inner feelings, which is significant in sentiment analysis and lie detection. Since micro-expression is expressed as subtle facial changes within particular facial action units, learning discriminative and generalized features for Micro-expression Recognition (MER) is challenging. To achieve the purpose, this paper proposes a novel MER framework that simultaneously integrates supervised Prototype-based Memory Contrastive Learning (PMCL) for discriminative feature mining and adds Self-expression Reconstruction (SER) as an auxiliary task and regularization for better generalization. In particular, the proposed SER module is forced as a regularization by reconstructing input ME from the randomly dropped patch-wise features in the bottleneck. And, the PMCL module globally compares historical and current cluster agents learned from training instances to enhance intra-class compactness and inter-class separability. Extensive experiments are conducted on three benchmarks, e.g., SMIC, CASME II, and SAMM, under evaluation criteria of both Composite Database Evaluation (CDE) and Single Database Evaluation (SDE) protocols. The results show our method surpasses other state-of-the-art approaches under various evaluation metrics, achieving overall 86.30% unweighed F1-score and 88.30% unweighed average recall on the composite dataset. Furthermore, the ablation studies verify the effectiveness of our SER for better generalization and PMCL for better discrimination in learning feature representation from limited micro-expression samples. [ABSTRACT FROM AUTHOR]

Long-tail distribution is a difficult challenge for knowledge graph embedding. We expect to solve the problem by complementing the information through the neighbor aggregation mechanism of GCN. However, the GCN method and its derivations are unable to learn the representation of edges. To address this problem, we propose RCGCN-TE, Relation Correlations-aware Graph Convolutional Network with Text-Enhanced for knowledge graph embedding, which is the first effort to enable GCN to learn the representation of relations directly. First, the pre-trained language model is used to extract semantic information. Then, the relation correlation graph is constructed by defining the relation relevance function based on the co-occurrence pattern and semantic similarity of relations. Finally, two GCNS are designed to learn entities and relations respectively. Experimental results on tasks such as triple classification and link prediction are better than the baseline. For example, Hits@10, Hits@3, and Hits@1 improved by 8.23 % , 37.49 % , and 46.94 % , respectively, on the entity prediction task. [ABSTRACT FROM AUTHOR]

Knowledge graph completion is one of the steps of knowledge graph construction. It aims to complete the incomplete triple data in the initially constructed knowledge graph and make the data in the knowledge graph more abundant and complete. At present, there are few studies on knowledge graph completion in the equipment field, and the attribute and relationship samples in the equipment knowledge graph are prone to uneven distribution, while the traditional knowledge graph completion method is difficult to solve the problem of uneven distribution of attribute and relationship samples. Therefore, this paper proposes an internal instance type completion method of equipment knowledge graph based on EP2TP-TRT. First, the TransE model is used to embed the relationship and attributes of the equipment instance, respectively. Then, the EP2TP model is used to map the instance attributes, and the TRT model is used to map the type relationship. Finally, the scores of the EP2TP and TRT models are integrated by designing different weights, and the training prediction is carried out to enhance the representation ability of instance information and type information. Compared with the mainstream advanced models, this method improves the MRR and HITS @ 1 indicators by about 0.89% and 2.1%, respectively. [ABSTRACT FROM AUTHOR]

Network anomaly detection plays a vital role in safeguarding network security. However, the existing network anomaly detection task is typically based on the one-class zero-positive scenario. This approach is susceptible to overfitting during the training process due to discrepancies in data distribution between the training set and the test set. This phenomenon is known as prediction drift. Additionally, the rarity of anomaly data, often masked by normal data, further complicates network anomaly detection. To address these challenges, we propose the PUNet network, which ingeniously combines the strengths of traditional machine learning and deep learning techniques for anomaly detection. Specifically, PUNet employs a reconstruction-based autoencoder to pre-train normal data, enabling the network to capture potential features and correlations within the data. Subsequently, PUNet integrates a sampling algorithm to construct a pseudo-label candidate set among the outliers based on the reconstruction loss of the samples. This approach effectively mitigates the prediction drift problem by incorporating abnormal samples. Furthermore, PUNet utilizes the CatBoost classifier for anomaly detection to tackle potential data imbalance issues within the candidate set. Extensive experimental evaluations demonstrate that PUNet effectively resolves the prediction drift and data imbalance problems, significantly outperforming competing methods. [ABSTRACT FROM AUTHOR]

Announced in November 2020, NTT's Medical and Health Vision states that the company will strive to put biological simulations using bio-digital twins (BDTs) into practical use to make effective use of medical resources, alleviate physical constraints of medical resources, provide continuous care from prevention to post-treatment, and provide precise care personalized for each individual. A BDT is a digital representation of a living organism, and the key to creating a BDT is how to express (quantify or symbolize) the functions of the living organism--and the physical and chemical mechanisms behind them--as digital information. We interviewed NTT Fellow Kunio Kashino of NTT Communication Science Laboratories, who has taken the challenge of creating a BDT, about the use of artificial intelligence in the biomedical field, stimulation and awareness in collaborative research across different disciplines, and his thoughts on the importance of striking a balance between what should be done and how it should be done. [ABSTRACT FROM AUTHOR]

The vast corpus of heterogeneous biomedical data stored in databases, ontologies, and terminologies presents a unique opportunity for drug design. Integrating and fusing these sources is essential to develop data representations that can be analyzed using artificial intelligence methods to generate novel drug candidates or hypotheses. Here, we propose Non-Negative Matrix Tri-Factorization as an invaluable tool for integrating and fusing data, as well as for representation learning. Additionally, we demonstrate how representations learned by Non-Negative Matrix Tri-Factorization can effectively be utilized by traditional artificial intelligence methods. While this approach is domain-agnostic and applicable to any field with vast amounts of structured and semi-structured data, we apply it specifically to computational pharmacology and drug repurposing. This field is poised to benefit significantly from artificial intelligence, particularly in personalized medicine. We conducted extensive experiments to evaluate the performance of the proposed method, yielding exciting results, particularly compared to traditional methods. Novel drug–target predictions have also been validated in the literature, further confirming their validity. Additionally, we tested our method to predict drug synergism, where constructing a classical matrix dataset is challenging. The method demonstrated great flexibility, suggesting its applicability to a wide range of tasks in drug design and discovery. [ABSTRACT FROM AUTHOR]

Anomaly detection has gained significant attention with the advancements in deep neural networks. Effective training requires both normal and anomalous data, but this often leads to a class imbalance, as anomalous data is scarce. Traditional augmentation methods struggle to maintain the correlation between anomalous patterns and their surroundings. To address this, we propose an adjacent augmentation technique that generates synthetic anomaly images, preserving object shapes while distorting contours to enhance correlation. Experimental results show that adjacent augmentation captures high-quality anomaly features, achieving superior AU-ROC and AU-PR scores compared to existing methods. Additionally, our technique produces synthetic normal images, aiding in learning detailed normal data features and reducing sensitivity to minor variations. Our framework considers all training images within a batch as positive pairs, pairing them with synthetic normal images as positive pairs and with synthetic anomaly images as negative pairs. This compensates for the lack of anomalous features and effectively distinguishes between normal and anomalous features, mitigating class imbalance. Using the ResNet50 network, our model achieved perfect AU-ROC and AU-PR scores of 100% in the bottle category of the MVTec-AD dataset. We are also investigating the relationship between anomalous pattern size and detection performance. [ABSTRACT FROM AUTHOR]

The cold-start problem in sequence recommendations presents a critical and challenging issue for portable sensing devices. Existing content-aware approaches often struggle to effectively distinguish the relative importance of content features and typically lack generalizability when processing new data. To address these limitations, we propose a content-aware few-shot meta-learning (CFSM) model to enhance the accuracy of cold-start sequence recommendations. Our model incorporates a double-tower network (DT-Net) that learns user and item representations through a meta-encoder and a mutual attention encoder, effectively mitigating the impact of noisy data on auxiliary information. By framing the cold-start problem as few-shot meta-learning, we employ a model-agnostic meta-optimization strategy to train the model across a variety of tasks during the meta-learning phase. Extensive experiments conducted on three real-world datasets—ShortVideos, MovieLens, and Book-Crossing—demonstrate the superiority of our model in cold-start recommendation scenarios. Compared to MetaCs-DNN, the second-best approach, CFSM, achieves improvements of 1.55%, 1.34%, and 2.42% under the AUC metric on the three datasets, respectively. [ABSTRACT FROM AUTHOR]

Multimodal sentiment analysis focuses on the fusion of multiple modalities. However, modality representation learning is a key step for better modality fusion, so how to fully learn the sentiment information of non-text modalities is a problem worth exploring. In addition, how to further improve the accuracy of sentiment polarity prediction is also a work to be studied. To solve the above problems, we propose a multimodal sentiment analysis model with effective context semantic modality fusion and sentiment polarity correction (CSMF-SPC). Firstly, we design a low-rank multimodal fusion network based on context semantic modality (CSM-LRMFN). CSM-LRMFN uses the bi-directional long short-term memory network to extract the context semantic features of non-text modalities, and the BERT to extract the features of text modality. Then, CSM-LRMFN adopts a low-rank multimodal fusion method to fully extract the interaction information among modalities with contextual semantics. Different from previous studies, to improve the accuracy of sentiment polarity prediction, we design a weight self-adjusting sentiment polarity penalty loss function, which makes the model learn more sentiment features that are conducive to model prediction through backpropagation. Finally, a series of comparative experiments are conducted on the CMU-MOSI and CMU-MOSEI datasets. Compared with the current representative models, CSMF-SPC achieves better experimental results. Among them, the Acc-2 (including zero) metric is increased by 1.41% and 1.58% on the word-aligned and unaligned CMU-MOSI datasets respectively; it is improved by 1.50% and 2.14% respectively on the CMU-MOSEI dataset, which indicates that the improvement of CSMF-SPC is effective. [ABSTRACT FROM AUTHOR]

Time series data are sequences of values that are obtained by sampling a signal at a fixed frequency, and time series classification algorithms distinguish time series into different categories. Among many time series classification algorithms, subseries-based algorithms have received widespread attention because of their high accuracy and low computational complexity. However, subseries-based algorithms consider the similarity of subseries only by shape and ignore semantic similarity. Therefore, the purpose of this paper is to determine how to solve the problem that subseries-based time series classification algorithms ignore the semantic similarity between subseries. To address this issue, we introduce the deep graph kernel technique to capture the semantic similarity between subseries. To verify the performance of the method, we test the proposed algorithm on publicly available datasets from the UCR repository and the experimental results prove that the deep graph kernel has an important role in enhancing the accuracy of the algorithm and that the proposed algorithm performs quite well in terms of accuracy and has a considerable advantage over other representative algorithms. [ABSTRACT FROM AUTHOR]

The large-size language model is able to implicitly extract informative semantic features from queries and candidate documents to achieve impressive reranking performance. However, the large model relies on its own large number of parameters to achieve it and it is not known exactly what semantic information has been learned. In this paper, we propose a multi-stage enhanced representation learning method based on Query-View (MERL) with Intra-query stage and Inter-query stage to guide the model to explicitly learn the semantic relationship between the query and documents. In the Intra-query training stage, a content-based contrastive learning module without considering the special token [CLS] of BERT is utilized to optimize the semantic similarity of query and relevant documents. In the Inter-query training stage, an entity-oriented masked query prediction for establish a semantic relation of query-document pairs and an Inter-query contrastive learning module for extracting similar matching pattern of query-relevant documents are employed. Extensive experiments on MS MARCO passage ranking and TREC DL datasets show that the MERL method obtain significant improvements with a low number of parameters compared to the baseline models. [ABSTRACT FROM AUTHOR]

Static node embedding algorithms applied to snapshots of real-world applications graphs are unable to capture their evolving process. As a result, the absence of information about the dynamics in these node representations can harm the accuracy and increase processing time of machine learning tasks related to these applications. Aiming at fill the gap regarding the inability of static methods to capture evolving processes on dynamic networks, we propose a biased random walk method named Evolving Node Embedding (EVNE). EVNE leverages the sequential relationship of graph snapshots by incorporating historic information when generating embeddings for the next snapshot. It learns node representations through a neural network, but differs from existing methods as it: (i) incorporates previously run walks at each step; (ii) starts the optimization of the current embedding from the parameters obtained in the previous iteration; and (iii) uses two time-varying parameters to regulate the behavior of the biased random walks over the process of graph exploration. Through a wide set of experiments we show that our approach generates better embeddings, outperforming baselines by up to 20% in a downstream node classification task. EVNE's embeddings achieve better performance than others, based on experiments with four classifiers and five datasets. In addition, we present seven variations of our model to show the impact of each of EVNE's mechanisms. [ABSTRACT FROM AUTHOR]

Nighttime dehazing is a challenging ill-posed problem due to the severe haze pollution and color attenuation. Since available daytime dehazing approaches cannot be consistently adapted to the nighttime case, this paper specifically designs representation learning to enhance an end-to-end nighttime dehazing network. To explore attentive and problem-driven representations, we propose to realize effective learning via knowledge distillation together with neural architecture search (NAS), which are implemented via two parallel branches of networks, namely, representation learning network (RLN) and guided dehazing network (GDN). The challenges in nighttime dehazing come from distilling the clear elements from the hazy representation and the restoration of real color from ambient illumination or low light condition. To this end, we introduce the subnetwork of RLN to learn the feature of the clear elements of the ground truth. We specifically introduce a NAS for exploring the color subspace of the latent embedding space of both RLN and GDN. With the guidance from the representation learned by RLN, GDN could well learn the clear elements from the hazy image and recover the real color of it effectively. Extensive experimental results on the synthetic and real-world datasets demonstrate the superiority of our method over the state-of-the-arts in nighttime dehazing. [ABSTRACT FROM AUTHOR]

Deep representation learning is a ubiquitous part of modern computer vision. While Euclidean space has been the de facto standard manifold for learning visual representations, hyperbolic space has recently gained rapid traction for learning in computer vision. Specifically, hyperbolic learning has shown a strong potential to embed hierarchical structures, learn from limited samples, quantify uncertainty, add robustness, limit error severity, and more. In this paper, we provide a categorization and in-depth overview of current literature on hyperbolic learning for computer vision. We research both supervised and unsupervised literature and identify three main research themes in each direction. We outline how hyperbolic learning is performed in all themes and discuss the main research problems that benefit from current advances in hyperbolic learning for computer vision. Moreover, we provide a high-level intuition behind hyperbolic geometry and outline open research questions to further advance research in this direction. [ABSTRACT FROM AUTHOR]

Copyright of Engineering Mechanics / Gongcheng Lixue is the property of Engineering Mechanics Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Imbalanced graph node classification is a highly relevant and challenging problem in many real-world applications. The inherent data scarcity, a central characteristic of this task, substantially limits the performance of neural classification models driven solely by data. Given the limited instances of relevant nodes and complex graph structures, current methods fail to capture the distinct characteristics of node attributes and graph patterns within the underrepresented classes. In this article, we propose REFUEL—a novel approach for highly imbalanced node classification problems in graphs. Whereas symbolic and neural methods have complementary strengths and weaknesses when applied to such problems, REFUEL combines the power of symbolic and neural learning in a novel neural rule-extraction architecture. REFUEL captures the class semantics in the automatically extracted rule vectors. Then, REFUEL augments the graph nodes with the extracted rules vectors and adopts a Graph Attention Network-based neural node embedding, enhancing the downstream neural node representation. Our evaluation confirms the effectiveness of the proposed REFUEL approach for three real-world datasets with different minority class sizes. REFUEL achieves at least a 4% point improvement in precision on the minority classes of 1.5–2% compared to the baselines. [ABSTRACT FROM AUTHOR]

The success of deep learning in computer vision and natural language processing communities can be attributed to the training of very deep neural networks with millions or billions of parameters, which can then be trained with massive amounts of data. However, a similar trend has largely eluded the training of deep reinforcement learning (RL) algorithms where larger networks do not lead to performance improvement. Previous work has shown that this is mostly due to instability during the training of deep RL agents when using larger networks. In this paper, we make an attempt to understand and address the training of larger networks for deep RL. We first show that naively increasing network capacity does not improve performance. Then, we propose a novel method that consists of (1) wider networks with DenseNet connection, (2) decoupling representation learning from the training of RL, and (3) a distributed training method to mitigate overfitting problems. Using this three-fold technique, we show that we can train very large networks that result in significant performance gains. We present several ablation studies to demonstrate the efficacy of the proposed method and some intuitive understanding of the reasons for performance gain. We show that our proposed method outperforms other baseline algorithms on several challenging locomotion tasks. [ABSTRACT FROM AUTHOR]

Entity alignment (EA) aims to automatically match entities in different knowledge graphs, which is beneficial to the development of knowledge-driven applications. Representation learning has powerful feature capture capability and it is widely used in the field of natural language processing. Compared with traditional EA methods, EA methods based on representation learning have better performance and efficiency. Hence, we summarize and analyze the representative EA approaches based on representation learning in this paper. We present the problem description and data preprocessing for EA and other related fundamental knowledge. We propose a new EA framework for the latest models, which includes information aggregation module, entity alignment module, and post-alignment module. Based on these three modules, the various technologies are described in detail. In the experimental part, we first explore the effect of EA direction on model performance. Then, we classify the models into different categories in terms of alignment inference strategy, noise filtering strategy, and whether additional information is utilized. To ensure fairness, we perform the comparative analysis of the performance of the models within the categories separately on different datasets. We investigate both unimodal and multimodal EA. Finally, we present future research perspectives based on the shortcomings of existing EA methods. [ABSTRACT FROM AUTHOR]

A human-like dialogue system should prioritize expressing empathy towards others, which entails two crucial aspects: (1) semantic cognition and (2) emotion detection. Previous approaches mainly model semantic and emotional dependencies by capturing emotion signal or leveraging external commonsense knowledge. However, they often ignore the syntactic information inherent in the utterances, which is crucial to understanding the subtle differences in the semantic and emotional meaning of words in different contexts. In this paper, we propose a novel framework for empathetic dialogue generation with Knowledge enhanced Dependency Tree (KnowDT). Specifically, the KnowDT captures relationships between token-level emotions by comprehensively considering of dependency tree and external knowledge. To incorporate syntactic and semantic information into context encoding, we improve the relation attention mechanism for a joint representation of adjacent nodes in dependency tree, and design a flexible tree positional encoding scheme for a broad variety of topologies without affecting their syntactic constituents. Furthermore, we present a level-ordered attention mechanism that extracts emotional features by encoding recursively along the topological structure of a dependency tree. Experimental results demonstrate that our model outperforms the state-of-the-art models in both automatic and human evaluation, in which the accuracy of emotion classification and the perplexity of generated responses are significantly improved. [ABSTRACT FROM AUTHOR]

Deep Cross-Modal Hashing (DCMH) has garnered significant attention in the field of cross-modal retrieval due to its advantages such as high computational efficiency and small storage space. However, existing DCMH methods still face certain limitations: (1) they neglect the correlation between labels, while label features exhibit high sparsity; (2) they lack fine-grained semantic alignment; (3) they fail to effectively address data imbalance. In order to tackle these issues, this paper introduces a framework named Semantic-Alignment Transformer and Adversary Hashing for Cross-modal Retrieval (SATAH). To the best of our knowledge, this is the first attempt at the Semantic-Alignment Transformer algorithm. Specifically, this paper first designs a label learning network that utilizes a crafted transformer module to extract label information, guiding adversarial learning and hash function learning accordingly. Subsequently, a Balanced Conditional Generative Adversarial Network (BCGAN) is constructed, marking the first instance of adversarial training guided by label information. Furthermore, a Weighted Semi-Hard Cosine Triplet Constraint is proposed to better ensure high-ranking similarity relationships among all items. Lastly, considering the correlation between labels, a semantic-alignment constraint is introduced to handle label correlation from a fine-grained perspective, capturing similarity on a global scale more effectively. Extensive experiments are conducted on multiple representative cross-modal datasets. In experiments with 64-bit hash code length, SATAH achieves average mAP values of 84.75%, 68.87%, and 68.73% on MIR Flickr, NUS-WIDE, and MS COCO datasets, respectively, outperforming state-of-the-art methods. The code is available at https://github.com/Daydaylight/SATAH. [ABSTRACT FROM AUTHOR]

Anomaly detection (AD) is a fundamental task in computer vision. It aims to identify incorrect image data patterns which deviate from the normal ones. Conventional methods generally address AD by preparing augmented negative samples to enforce self-supervised learning. However, these techniques typically do not consider semantics during augmentation, leading to the generation of unrealistic or invalid negative samples. Consequently, the feature extraction network can be hindered from embedding critical features. In this study, inspired by visual attention learning approaches, we propose CutSwap, which leverages saliency guidance to incorporate semantic cues for augmentation. Specifically, we first employ LayerCAM to extract multilevel image features as saliency maps and then perform clustering to obtain multiple centroids. To fully exploit saliency guidance, on each map, we select a pixel pair from the cluster with the highest centroid saliency to form a patch pair. Such a patch pair includes highly similar context information with dense semantic correlations. The resulting negative sample is created by swapping the locations of the patch pair. Compared to prior augmentation methods, CutSwap generates more subtle yet realistic negative samples to facilitate quality feature learning. Extensive experimental and ablative evaluations demonstrate that our method achieves state-of-the-art AD performance on two mainstream AD benchmark datasets. [ABSTRACT FROM AUTHOR]

In fashion e-commerce, predicting item compatibility using visual features remains a significant challenge. Current recommendation systems often struggle to incorporate high-dimensional visual data into graph-based learning models effectively. This limitation presents a substantial opportunity to enhance the precision and effectiveness of fashion recommendations. In this paper, we present the Visual-aware Graph Convolutional Network (VAGCN). This novel framework helps improve how visual features can be incorporated into graph-based learning systems for fashion item compatibility predictions. The VAGCN framework employs a deep-stacked autoencoder to convert the input image's high-dimensional raw CNN visual features into more manageable low-dimensional representations. In addition to improving feature representation, the GCN can also reason more intelligently about predictions, which would not be possible without this compression. The GCN encoder processes nodes in the graph to capture structural and feature correlation. Following the GCN encoder, the refined embeddings are input to a multi-layer perceptron (MLP) to calculate compatibility scores. The approach extends to using neighborhood information only during the testing phase to help with training efficiency and generalizability in practical scenarios, a key characteristic of our model. By leveraging its ability to capture latent visual features and neighborhood-based learning, VAGCN thoroughly investigates item compatibility across various categories. This method significantly improves predictive accuracy, consistently outperforming existing benchmarks. These contributions tackle significant scalability and computational efficiency challenges, showcasing the potential transformation of recommendation systems through enhanced feature representation, paving the way for further innovations in the fashion domain. [ABSTRACT FROM AUTHOR]

Due to the heterogeneity gap in multi-view data, researchers have been attempting to apply these data to learn a co-latent representation to bridge this gap. However, multi-view representation learning still confronts two challenges: (1) it is hard to simultaneously consider the performance of downstream tasks and the interpretability and transparency of the network; (2) it fails to learn representations that accurately describe the class boundaries of downstream tasks. To overcome these limitations, we propose an interpretable representation learning framework, named interpretable multi-view proximity representation learning network. On the one hand, the proposed network is customized by an explicitly designed optimization objective that enables it to learn semantic co-latent representations while maintaining the interpretability and transparency of the network from the design level. On the other hand, the designed multi-view proximity representation learning objective function encourages its learned co-latent representations to form intuitive class boundaries by increasing the inter-class distance and decreasing the intra-class distance. Driven by a flexible downstream task loss, the learned co-latent representation can adapt to various multi-view scenarios and has been shown to be effective in experiments. As a result, this work provides a feasible solution to a generalized multi-view representation learning framework and is expected to accelerate the research and exploration in this field. [ABSTRACT FROM AUTHOR]

Self-supervised learning has made significant progress in point cloud processing. Currently, the primary tasks of self-supervised learning, which include point cloud reconstruction and representation learning, are trained separately due to their structural differences. This separation inevitably leads to increased training costs and neglects the potential for mutual assistance between tasks. In this paper, a self-supervised method named PointUR-RL is introduced, which integrates point cloud reconstruction and representation learning. The method features two key components: a variable masked autoencoder (VMAE) and contrastive learning (CL). The VMAE is capable of processing input point cloud blocks with varying masking ratios, ensuring seamless adaptation to both tasks. Furthermore, CL is utilized to enhance the representation learning capabilities and improve the separability of the learned representations. Experimental results confirm the effectiveness of the method in training and its strong generalization ability for downstream tasks. Notably, high-accuracy classification and high-quality reconstruction have been achieved with the public datasets ModelNet and ShapeNet, with competitive results also obtained with the ScanObjectNN real-world dataset. [ABSTRACT FROM AUTHOR]

The purpose of representation learning is to encode the entities and relations in a knowledge graph as low-dimensional and real-valued vectors through machine learning technology. Traditional representation learning methods like TransE, a method which models relationships by interpreting them as translations operating on the low-dimensional embeddings of a graph's entities, are effective for learning the embeddings of knowledge bases, but struggle to effectively model complex relations like one-to-many, many-to-one, and many-to-many. To overcome the above issues, we introduce a new method for knowledge representation, reasoning, and completion based on multi-translation principles and TransE (TransE-MTP). By defining multiple translation principles (MTPs) for different relation types, such as one-to-one and complex relations like one-to-many, many-to-one, and many-to-many, and combining MTPs with a typical translating-based model for modeling multi-relational data (TransE), the proposed method, TransE-MTP, ensures that multiple optimization objectives can be targeted and optimized during training on complex relations, thereby providing superior prediction performance. We implement a prototype of TransE-MTP to demonstrate its effectiveness at link prediction and triplet classification on two prominent knowledge graph datasets: Freebase and Wordnet. Our experimental results show that the proposed method enhanced the performance of both TransE and knowledge graph embedding by translating on hyperplanes (TransH), which confirms its effectiveness and competitiveness. [ABSTRACT FROM AUTHOR]

Chemical engineers heavily rely on precise knowledge of physicochemical properties to model chemical processes. Despite the growing popularity of deep learning, it is only rarely applied for property prediction due to data scarcity and limited accuracy for compounds in industrially-relevant areas of the chemical space. Herein, we present a geometric deep learning framework for predicting gas- and liquid-phase properties based on novel quantum chemical datasets comprising 124,000 molecules. Our findings reveal that the necessity for quantum-chemical information in deep learning models varies significantly depending on the modeled physicochemical property. Specifically, our top-performing geometric model meets the most stringent criteria for "chemically accurate" thermochemistry predictions. We also show that by carefully selecting the appropriate model featurization and evaluating prediction uncertainties, the reliability of the predictions can be strongly enhanced. These insights represent a crucial step towards establishing deep learning as the standard property prediction workflow in both industry and academia. Scientific contribution We propose a flexible property prediction tool that can handle two-dimensional and three-dimensional molecular information. A thermochemistry prediction methodology that achieves high-level quantum chemistry accuracy for a broad application range is presented. Trained deep learning models and large novel molecular databases of real-world molecules are provided to offer a directly usable and fast property prediction solution to practitioners. [ABSTRACT FROM AUTHOR]

Brain networks based on functional magnetic resonance imaging (fMRI) provide a crucial perspective for diagnosing brain diseases. Representation learning has recently attracted tremendous attention due to its strong representation capability, which can be naturally applied to brain disease analysis. However, traditional representation learning only considers direct and local node interactions in original brain networks, posing challenges in constructing higher-order brain networks to represent indirect and extensive node interactions. To address this problem, we propose the Continuous Dictionary of Nodes model and Bilinear-Diffusion (CDON-BD) network for brain disease analysis. The CDON model is innovatively used to learn the original brain network, with its encoder weights directly regarded as latent features. To fully integrate latent features, we further utilize Bilinear Pooling to construct higher-order brain networks. The Diffusion Module is designed to capture extensive node interactions in higher-order brain networks. Compared to state-of-the-art methods, CDON-BD demonstrates competitive classification performance on two real datasets. Moreover, the higher-order representations learned by our method reveal brain regions relevant to the diseases, contributing to a better understanding of the pathology of brain diseases. [ABSTRACT FROM AUTHOR]

Speech assistive system for people with neuro disorders is a highly challenging task till date. Any kind of neuro cognitive disability affects the speech production mechanism that leads to speech impairment. Representation learning methods have recently emerged to improve the outcome of machine learning algorithms. In case of complex recognition tasks such as disordered speech recognition, learning compact and efficient representations for disordered speech utterances is important. Recent deep learning-based architectures need sufficiently large amount of impaired speech samples which are tedious with respect to neurologically disabled people. In this work, we focus on proposing a representation learning approach that uses traditional sequential model such as Hidden Markov Model (HMM) which works moderately well with small amount of impaired speech data. We propose a novel sequence to vector-based HMM State Sequence (HMM-SS) approach which is very compact and has proved to be an efficient representation for disordered speech utterances. The efficiency of the proposed HMM-SS approach is assessed using four datasets, namely 50 words of TORGO, 100-common words dataset of the UA-SPEECH, 50 help-seeking words and 100-common words of Impaired speech corpus in Tamil corpus. The proposed approach outperforms the baseline HMM, DNN-HMM and a recent state-of-the-art approach for all four datasets. The discriminative ability and the compactness of the proposed representation proved effective for disordered speech recognition. [ABSTRACT FROM AUTHOR]

The architecture of the brain is too complex to be intuitively surveyable without the use of compressed representations that project its variation into a compact, navigable space. The task is especially challenging with high‐dimensional data, such as gene expression, where the joint complexity of anatomical and transcriptional patterns demands maximum compression. The established practice is to use standard principal component analysis (PCA), whose computational felicity is offset by limited expressivity, especially at great compression ratios. Employing whole‐brain, voxel‐wise Allen Brain Atlas transcription data, here we systematically compare compressed representations based on the most widely supported linear and non‐linear methods—PCA, kernel PCA, non‐negative matrix factorisation (NMF), t‐stochastic neighbour embedding (t‐SNE), uniform manifold approximation and projection (UMAP), and deep auto‐encoding—quantifying reconstruction fidelity, anatomical coherence, and predictive utility across signalling, microstructural, and metabolic targets, drawn from large‐scale open‐source MRI and PET data. We show that deep auto‐encoders yield superior representations across all metrics of performance and target domains, supporting their use as the reference standard for representing transcription patterns in the human brain. [ABSTRACT FROM AUTHOR]

Self-supervised representation learning and deep clustering are mutually beneficial to learn high-quality representations and cluster data simultaneously in centralized settings. However, it is not always feasible to gather large amounts of data at a central entity, considering data privacy requirements and computational resources. Federated Learning (FL) has been developed successfully to aggregate a global model while training on distributed local data, respecting the data privacy of edge devices. However, most FL research effort focuses on supervised learning algorithms. A fully unsupervised federated clustering scheme has not been considered in the existing literature. We present federated momentum contrastive clustering (FedMCC), a generic federated clustering framework that can not only cluster data automatically but also extract discriminative representations training from distributed local data over multiple users. In FedMCC, we demonstrate a two-stage federated learning paradigm where the first stage aims to learn differentiable instance embeddings and the second stage accounts for clustering data automatically. The experimental results show that FedMCC not only achieves superior clustering performance but also outperforms several existing federated self-supervised methods for linear evaluation and semi-supervised learning tasks. Additionally, FedMCC can easily be adapted to ordinary centralized clustering through what we call momentum contrastive clustering (MCC). We show that MCC achieves state-of-the-art clustering accuracy results in certain datasets such as STL-10 and ImageNet-10. We also present a method to reduce the memory footprint of our clustering schemes. [ABSTRACT FROM AUTHOR]

As lovely as bunnies are, your sketched version would probably not do them justice (Fig. 1). This paper recognises this very problem and studies sketch quality assessment for the first time—letting you find these badly drawn ones. Our key discovery lies in exploiting the magnitude ( L 2 norm) of a sketch feature as a quantitative quality metric. We propose Geometry-Aware Classification Layer (GACL), a generic method that makes feature-magnitude-as-quality-metric possible and importantly does it without the need for specific quality annotations from humans. GACL sees feature magnitude and recognisability learning as a dual task, which can be simultaneously optimised under a neat cross-entropy classification loss with theoretic guarantee. This gives GACL a nice geometric interpretation (the better the quality, the easier the recognition), and makes it agnostic to both network architecture changes and the underlying sketch representation. Through a large scale human study of 160,000 trials, we confirm the agreement between our GACL-induced metric and human quality perception. We further demonstrate how such a quality assessment capability can for the first time enable three practical sketch applications. Interestingly, we show GACL not only works on abstract visual representations such as sketch but also extends well to natural images on the problem of image quality assessment (IQA). Last but not least, we spell out the general properties of GACL as general-purpose data re-weighting strategy and demonstrate its applications in vertical problems such as noisy label cleansing. Code will be made publicly available at https://github.com/yanglan0225/SketchX-Quantifying-Sketch-Quality. [ABSTRACT FROM AUTHOR]